To know whether something is bullish or bearish, we need to know not only its value to customers, but also its ability to differentiate itself from competitors

To know whether something is bullish or bearish, we need to know not only its value to customers, but also its ability to differentiate itself from competitors and whether it will be able to maintain the value margin. One clear value that OpenAi can offer customers right now is the ability to do inference at marginal cost. Even when open source models get close to the quality of ChatGPT, they will be expensive to run, let alone to fine-tune or update. Knowledge is a constantly evolving thing. If you don't have a big budget, it will be very hard to keep up. Unless we can find a way for the community to pool the resources to train open source models and keep them up to date. That might be a bit different for big companies where they want to run their own models. The other factor is that Google and others are certainly not going to sit still. There is no reason to believe that someone as resourceful as Google cannot come up with something as good as ChatGPT, if not better. Companies like Meta are playing the open source card, so they will be the first to benefit directly from the community. So the market will change, but dramatically. It's far too early to bet on any of them (or none of them). My approach is to diversify, wait and see.